The Progressive Corporation (PGR) Q4 2016 Earnings Report, Transcript and Summary

Welcome to the Progressive Corporation's Investor Relations Conference Call. This conference call is also available by an audio webcast. [Operator Instructions] In addition, this conference is being recorded at the request of Progressive. If you have any objections, you may disconnect at this time. The company will not make detailed comments in addition to those provided in its annual report on Form 10-K and the letter to shareholders, which have been posted to the company's website, and we'll use this conference call to respond to questions. Acting as moderator for the call will be Julia Hornack. At this time, I will turn the call over to Ms. Hornack.

Thank you, Martia. Good morning. Welcome to Progressive's conference call. Participating on today's call are Tricia Griffith, our CEO; John Sauerland, our CFO; and Bill Cody, our Chief Investment Officer. The call is scheduled to last about an hour. As always, our discussions on this call may include forward-looking statements. These forward-looking statements are based on management's current expectations and are subject to many risks and uncertainties that could cause actual events and results to differ materially from those discussed during this call. Additional information concerning those risks and uncertainties is available in our 2016 annual report on Form 10-K, where you will find discussions of the risk factors affecting our businesses, safe harbor statements related to forward-looking statements and other discussions of the risks, uncertainties and other challenges we face. These documents can be found via the Investors page of our website, progressive.com. Martia, we are now ready to take our first question.

[Operator Instructions] The first question will be coming from Kai Pan from Morgan Stanley.

First question for Tricia. We have seen rising lost cost across the auto industry and competitor are raising price, some aggressively, which creates some market opportunity. At the same time, you are close to your 96% target and pulled back on advertising in the second half. Would you continue to pull back on marketing to improve your margin? What -- would you seize the new -- sort of this, market dislocation to grow your business even at the risk of running -- breaching the 96% target?

Thanks, Kai. We did that last year. As you know, we have a very clear objective to grow as fast as we can and make at least $0.04 of underwriting profit. And we got close to that last year towards the latter half. As you know, about 2.5 points that were catastrophes, and so we pulled back a little bit. We are fully back into our advertising mode. And we're obviously seeing some of our competitors and they've acknowledged that they need a lot of rate in the system. We added about 5.5 points last year. We feel very good about our rate accuracy. We feel great about being able to turn on our advertising. And we really owe it to our owners to consistently talk about our objective of growing as fast as we can at a 96 combined ratio. So we don't take that lightly. Our decision to pull back we knew would have some short-term opportunities loss, but we were prepared for that because of how important it is for our owners to know that we'll do whatever we can to make sure we reach our objectives. So I mean, this year so far, as you saw our January results, we feel great where we're at. Our future trend looks to be about 4%. Obviously, we have to react if that changes but we feel like we made the appropriate decision to pull back, and we're excited to be back in with full advertising.

And separately, we have seen traffic fatality has risen about 40,000 last year and the car manufacturer are actually adopting the sort of like advanced technology, some ADAS system. Do you actively giving discount to cars with ADAS? And how do you think of the tech adoption in cars could impact both the frequency and the severity in both near term as well as over the long run?

Yes, we have seen, obviously, a lot in the news regarding fatalities and a lot regarding autonomous vehicles and reasoning for that. We clearly have a lot of data based on our Snapshot model. We have almost 19 billion miles and over 2 million usage-based PIP. So we look at that closely. And we have seen that mileage has gone up for participant. In fact, in the last 26 months it's gone up, mostly based on longer trips. So we're looking at that. And we're also looking at -- from an R&D perspective, to be able to look at those emerging technologies. And if we feel like they correlate to less losses, putting that into our product both now and future. So yes, that's a big part of our R&D. We really do, though, believe the future -- when we think about the future, we want to have data that correlates to all the things you hear and read about and have a basis on what that means as far as loss trends.

Thanks, Kai. Martia, we'll now take the next question.

The next question comes from the line of Elyse Greenspan from Wells Fargo.

My first question is on the prospect for corporate tax reform in the U.S. As you guys think about lower taxes, do you think that, that would fall to the bottom line? Or would you potentially be forced by regulators to push that through to your insurance? And if you were, I guess, forced to push through the lower tax rate, and then, how do you see this playing out over what kind of timeframe it turns out the prices you're able to charge?

I'll take that one. This is John Sauerland. We don't have a consensus view on tax reform. And certainly, we're not in a position to say what we would do, specifically, as there's no specific changes to react to at this juncture. So generally speaking, however, I would say our inclination would be to allow any tax benefits, presuming there would be any to fall to the bottom line. So we continue to focus on the 96, as Tricia just mentioned. I would expect we would continue to do that. You mentioned regulators choosing to force that change through to consumers. We have not seen that previously. States charge premium taxes, which are a little less transparent to you all, and that they show up actually in our acquisition costs. So we pay between $350 million and $400 million annually in state premium tax in lieu, generally speaking, of state income taxes. And that's generally where the regulators, who are state-based, would focus in terms of allowing taxes to flow through or not to our premiums. So we don't have a consensus view whether to be a change. Generally speaking, I think it's fair to say, we would allow that to fall to the bottom line.

Yes.

And my second question, as we think about 2016, do you -- can you point to or just kind of talk to the loss picks that you use? And if maybe there was some element of conservatism that you guys have built in, I guess, just as we think about some of the missteps and some of your competitors that obviously you know have missed some of the rising trends, especially on the severity side. If you can just kind of talk us through how you set your picks for 2016.

Yes, I mean, we're obviously using the term setting a pick is broad because we have our objective. But we look at every product, new and renewal, and we look at loss trends based on the past and what we're seeing pretty early on with our usage-based information. And we set them accordingly and to make sure that we meet or beat our targets. And those targets are very different per state, new and renewal, product, customer. And so to say one pick, it kind of simplifies it. We really dig deep in the segmentation and then ultimately build that into our product model.

To that I would add, as Tricia noted, we did take 5.5% rate increases in our auto programs last year, which is above what we normally do in a year. Additionally, we added over 9 points of rate to our Commercial Lines products. So we have been reacting to trends fairly aggressively, and I think that's why you've seen our combined ratios where they're at.

Yes, one more thing on that. I think -- John brings up the commercial product. We early -- we made early and decisive decisions on that and dug in. Even though in June and July when we first saw frequency increase a little bit, that was -- we thought that was -- that could have been a seasonality issue. But we dug into it and we were able to react, starting in August, to get that 9%. So I think part of it is, things do pop up I think on any insurance company unknown. So when you do, it's how you react. And our ability to react swiftly has always been one of our strengths.

We've also reacted with underwriting changes. And so especially in our commercial vehicle, when we see segments for which we are not adequately priced, we can restrict them very rapidly. And we are also doing that now in our personalized programs pretty robustly. We have a number of leverage we can pull to either take a second look at business before we commit to it or completely reject business if we are pretty confident that, that business is going to run above our targets.

Thanks, Elyse. Martia, please, the next caller?

The next question comes from the line of Josh Shanker.

So following up on Elyse's question. If I turn on the TV right now, someone's going to tell me to give them 15 minutes and save me $500 more or 15% more on auto insurance. Following up on Elyse's question on tax, what have you learned about the competitive environment in this industry? And if a bunch of profitable auto insurers do get a tax break, how is that going to affect advertising and the behaviors of the industry?

I'll start with that. I'll have John talk a little a bit more about tax. And I think John was clear on any tax reduction we believe would benefit us, and we would want that to go to the bottom line through lower prices for our customers and then everyone wins. For us, the advertising is really about what you think you can get, harvest in terms of new business. And you can listen to 5 different commercials and everyone is going to save you $500 and all of them are incorrect at some point. But we believe the advertising is important to be able to be top of mind, be short on the list in terms of when you make that call, when you get possibly rate shock as more and more customers are likely to take greater rates at a higher pace. So to me, those are a little bit separate from the tax issue, which will -- which that will be an answer that we get some time in the future. For us, it's really about how much we want to advertise at the right cost, so at our targeted acquisition cost, to get the right customers in the door that we want.

And I'll just add to that. So the taxes that we include in our pricing indications are the premium taxes, the state level taxes. The federal taxes, we have been on the high side, frankly, in terms of what we pay. You can think of our effective tax rate over time as closer to 32%, 33%. We have taken some actions more recently to be proactive in managing that. I would say we think that our premium tax management is some of the best in the industry. And again, that's 1 out to 2 points, that is in our pricing indications, is in the expense ratio we report every month. But we haven't been extremely proactive with our federal tax rate. We're starting to take actions to do so. We did 2 investment tax credit deals last year that moved our effective rate down about 4 points. That doesn't all fall to the bottom line, but we do get some benefits from that, and I would expect us to continue to do so. But our 96, if you followed us for us a while and as Tricia just said, is sort of chiseled in stone for us. And I don't expect we'll change that.

Okay. And there is -- just one more in. Publicly, the available information isn't great, but I have a sense about where you've been raising prices on publicly available information. And it seems like in your disclosure in the SEC financials, it seems to be that the premium per policy is up in excess of the rate filings that I'm looking at. I know a lot of competitors say there's a leakage sometimes, that if we file, sometimes customers buy down and therefore, we don't get the complete benefit of a rate increase. But I've never seen a excess of rate increase come through. I'm trying to understand the mechanism that might be causing that to happen.

Sure, that's a great question. And if you take rates up or you take rates down, you won't see exactly in the reported average premium what was filed in the rate change. A lot of stuff is going on in countrywide. So we have a lot of products. We have a lot of states that have a higher and lower average premiums. We grow more or less in certain segments because of our product changes. So our product changes are not just base rate increases, if you will. They're more frequently targeted at segments in which we think we can lower rates and get more competitive and still hit our 96 or if we need to raise rates to hit our 96. So you see what we think of as mix shifts in book, across our product, across states. And so the reported average premium change will never match the reported rate change filed with the departments. And you're correct. Our average premiums have been going up at a solid clip. And actually, in the auto program is closer to what we have filed in the commercial programs. We have had very large mix shifts to higher average premium segments over time. So if you look back, especially over the past 2 years, you'll see average premium changes going up a lot in Commercial Lines when we weren't even raising rates. More recently, we are raising rates, but you wouldn't even see that yet in the average premiums because that takes a while to flow through the book as we generally have reported our auto programs on 6-month policies generally and our Commercial Lines are on the annual. So yes, we've enjoyed average premium gains, which is fantastic. They won't necessarily match the filed rate changes.

Thanks, Josh. Martia, we'll take the next caller, please.

Your next question comes from the line of Ryan Tunis from Crédit Suisse.

This is Crystal Lu in for Ryan. My question is on the new property reinsurance cover. Can you walk us through your thinking behind the change from the aggregates of your storm cover to the new stop-loss reinsurance agreement that caps the losses excluding the named storms at a 63 loss ratio? And can you also give some color on the cost of this type of reinsurance and the amount of premium that's tied to it?

Yes, I'm going to have Trevor, the CFO from ASI, take that one.

Good morning. The stop loss -- to do that, it was a simple consideration of getting additional coverage for a reasonable cost. While the aggregate stop loss doesn't include named storms and liability, it includes other parallels, such as winter storm, wildfire and earthquake. The previous aggregate was limited to just severe thunderstorms. For the cost, it's difficult to say what our total reinsurance cost will be since a large portion of our program is placed in June. But we believe the reinsurance market remain soft. And as we expand our footprint to states that are not coastal, we're hopeful that our direct premiums will increase at a faster rate than our seeded premiums.

Thanks, Crystal. Martia, we'll take the next caller, please.

The next question comes from the line of Bob Glasspiegel from Janney Montgomery.

And I'm going to follow up on that question with property. Tricia, now that we've had a decent bit of time post the acquisition, how would you assess where you are both with respect to growth and profitability in the property book?

Yes, thanks, Bob. We feel good about where we're at. Obviously, we've talked a lot about the Platinum program, which was the big reason that we started this acquisition or began our partnership with ASI to have access to an addressable market we didn't have, especially on the agency side. So we feel good about growth. We feel good about the targets that we've set. Obviously, last year was more volatile on the property side from the CAT perspective, and that's one of the reasons why we're always looking at reinsurance and how that'll work, especially as we expand in more and more states of ASI. So as of last year, we were in 39 because we added a state in the last couple of months. But we continue to feel really good. And obviously, as we look at each state and at both products, and we talked about this a little bit in October, we look at are we competitive on the preferred auto? Are we competitive on the home? Are we competitive in both? And so when we say no to any of those, we do a deep dive with our R&D department in ASIs to understand how can we make sure we have broad enough coverage to reach our goals. Again, the 96, that property is a component of that, and to grow as much of that preferred business as we can. And so we're doing that obviously through our Platinum agents. We have specific sales reps that call on those agents to make sure they reach the goals we set for them. And we'll double the amount of the agents in 2017. Clearly, we obviously would love to grow more and more Robinsons. We are proud of the fact that we were able to grow our PIP base over 50% last year. But again, it's a on a small base, so I don't want to say that we -- there's a lot of runway on there, Bob. And we know -- I think one of the things we learned from this is that changing agents' behavior is slower than we thought. But we are bullish on where we're going with that product, with the growth and with the profitability in all of our products.

A follow-up is Commercial Auto. I think you came in around a 93 last year and your full last 2 months have been, very good profitability month. Are we now in a position to grow? Or do you need more data to make sure that your empty rounds [ph] around the VIX [ph] which you started in August, as you said? Are we in a position to grow in that business this year? I'm sorry.

Yes, we feel, I'm sorry. Yes, that's okay. We feel -- obviously, we're in a much better position. We think of Q3 last year coming in around a 99 and then Q4 coming in around a 91. We knew we were on the right trajectory. But John -- like John said, we had the 9% rate. It takes a little bit longer to earn in based on annual policies. And we had a fair amount of underwriting restrictions on new business in those high-frequency segments. What we're doing as we have more and more knowledge -- and over the next couple of months we'll determine if we can remove some of those restrictions. And that will be -- first and foremost, what we will look at in terms of to spur on new growth. And in addition, frequency is still a little bit elevated. We think some of that's seasonality and some of it's just based on miles driven and other frequency triggers. So we're going to keep our eye on it closely. But we will be assessing each month whether we should remove some or all of those underwriting restrictions for new business that we put into place.

Thanks, Bob. Martia, please, the next caller.

The next question comes from Meyer Shields from KBW.

Tricia, you talked about the PIF growth in the Robinsons segment. And I'm wondering if you can just clarify, when you price that business, are you pricing for a 96 during the initial calendar year? Or is it a 96 over the lifetime of individual policies?

We price different segments. We don't talk publicly about how we price each segment, but we -- obviously, on an aggregate calendar year, we want to get to a 96. And on a cohort pricing for lifetime, 96. So we look at both of those and determine on incoming business exactly how we can do that. And obviously, it's different in new and renewal and different in the channels.

Okay. Understood. A separate question, there's obviously tremendous uncertainty about the Affordable Care Act and what may or may not replace it. Can you talk about how you incorporate that in some of your medium tail, I'm thinking BI lines?

Well, we like certainty when we think of trends. Obviously, we watch those and we have an area of our company in the R&D section that thinks about trends and thinks about runway. And we model out different things that can change and affect medical costs as well as other costs, whether sheet metal costs. And so, we look at those but we really can't react until we have data that we see or are reasonably certain of. So those things are so up in the air. We really are listening, reading, modeling out. But we'll react when we were able to see data comes through the system or when we have reasonable knowledge of what we believe will happen.

Okay. And by data, do you mean actual losses? Or output from the R&D department?

Losses.

Thanks, Meyer. Martia, please, the next caller.

The next question comes from Ian Gutterman from Balyasny.

I guess I wanted to start on the PIF growth side. I had noticed something interesting, I guess, looking forward here. We're about to hit a point where it looks like you're going to cross over where agency PIF will be growing the same or even faster than direct PIF as direct starts to slow and agency keeps accelerating. And then that's a pretty rare occurrence, at least in all the history I see in -- from the company. Just curious on your observations on that. Is that something you guys are expecting? And going forward, do you think agency can keep up with direct? Or are we just sort of in the sweet spot for agency and direct's -- as you said, slow that advertising in such that -- we shouldn't expect that to continue? I'm just kind of curious how you're looking at the 2 channels.

Yes, I mean, obviously, we want to grow as much as we can in both channels, and that's the formula we want. We've been proud of the agency growth, it was long awaited, and we don't want that to stop. And clearly, when you pull up back on advertising, it's going to have a much bigger impact on the direct side. So that will start to come as the year progresses because we're fully back into the -- and into our advertising that we budgeted for. So to say we want to plan to grow one more than the other, we want to grow both substantially. That's one of the reasons why we have the acquisition of ASI because we really saw that addressable market, mostly on the agency side to be able to have this customer, the Robinson auto and home bundle and have access to that. So we think there's a lot of room to grow on the agency channel. And clearly, we want to be available when, where and how customers want to buy. So if it's on the agency channel, direct, through mobile, however that is, we are going to try to fully throttle on every channel in every way we can get good and profitable business.

Got it. That's helpful. And then just a follow-up on John's earlier question about premium growth versus PIF. And you talked about some of the mix impacts from what states are growing and such. But I assume part of it is also, as you're bundling more, as you're getting more higher-end customers who just buy more insurance and things like that. Is there a way to help us? I think I've asked this in the past but maybe it's just for an update on it. Just to help us understand a little bit better inside that mix shift? And A, how to think about what is sort of a normal level of positive mix is on top of PIF and average premium? And then, is there anything -- I assume the losses for the most part, maybe there's obviously, some bucks are a little better than others, but I assume the losses are largely going to be variable with the growth. But does expense leverage come at off from shifting the mix? Or it is expense leverage are really more about unit growth?

I may not hit all of these. So ask if I don't and John, weigh in if I didn't. So we're going to look at, obviously -- PIF growth is the relationship of new and renewal. So the new business comes in through more advertising, having a competitive product on the market and the retention of something I talked about in our letter regarding just being able to have the right rate, have the ability to -- have the right rate at the right time and also nurture our customers. So when we look at growth, we really do try to focus on unit growth. And the premium comes with a lot of different mixes. So it could be state mix. They vary greatly in state. Obviously, full coverage, which more of the Robinsons will have full coverage, so that affects premium. So those shifts, I think come as our shifts in having more Robinsons come, but also just the trends that we're seeing from loss cost. So it's -- there's a lot of inputs. We focus on unit growth and obviously, retention's a big piece that we've been increasingly -- not necessarily just paying more attention to, but really having a lot of focus on. So expenses, that was the last thing I think you talked about. That matters a lot. We see the companies that grow and grow profitably have a -- care a lot about cost. We always do. We continue to do that. Only because if you can be more efficient and do more today than you did yesterday with the same amount of people, we're able to bake that into our prices and we're more competitive. So that's how we grow. But unit growth continues to be our mantra and with all those caveats I just said. Yes.

Yes. The only thing I'd add to that is in terms of expense management, we focus a lot on what we call our nonacquisition expense ratio. If you look at that 20 expense ratio, you can think about half of it is, what we call acquisition expenses. Those are either commission or media spend dollars. And as we desire to continue to grow, those costs are likely to continue to rise. Commission, obviously, along with premium. Advertising cost, we want to be as efficient as possible. But more spend that we can deploy to our advertising at our targets, we certainly want to do as much of that as we can. But on the nonacquisition expense side, we think of those as sort of structural costs that we want to work to continue to reduce. If you look over the past maybe 7 to 8 years, we've probably taken 2 points out of our nonacquisition expense ratio. We'd like to take more out of that. And that comes from process improvements getting more efficient, moving customers more toward self-service. But it is largely driven by people. So as Tricia said, we want to do more if we can with the same amount of people. At the same time, more recently, we've been adding a ton of employees. We want to make sure we are adequately staffed as we grow. It's important when you're adding a lot of people to ensure that you're not understaffed as we have a lot of employees now with tenure with the company, less than a year and that are handling claims. So you want to make sure you're adequately staffed when you're growing a lot. So that's kind of the place we're in right now. But our long-term objective is to continue to reduce our nonacquisition expense ratio.

Yes. No, on the LAE side as well. So our whole team is focused on what we can do in the next several years, really, chipping away at that, and -- but like John said, you have to invest in both people and technology. And you read that hopefully in the K, how we've invested a lot in quoting systems and other systems. But for us, it's really about investing for the future in order to be more efficient. And -- but we don't want to shortcut that. When I talk about 96 growth as fast as you can, the only caveat to that is if we're able to service our customers in the way that we know we need to both in the claims side and the CRM side.

Absolutely. And that's why John, I was just focusing on that, that nonacquisition expense, thinking about sort of the 2 things that brought up and trying to think what the trade-offs are. Meaning, one is, just if a, Sam, I guess, generates $800 of premium to, as an example, in a Robinson across 2.5 cars, let's say is $3,000, is there more leverage? But then again, does it cost more to serve them because you have to invest for the growth? That's why I was trying to figure out that trade-off.

Yes, we look at all of that. And what we've made very clear is that we love Sam as long as we can make our target margin on Sam. But we also look at that from a CRM side to determine -- if someone's calling in a lot, are there ways that we can get information out proactively to the people that need those. So we have -- I think we talked about that a little bit in the October meeting. Just the -- all of the data around the convection [ph] and needs and understanding proactively what each customer needs. And Sam's calling in all the time for the same question. How can we do that more efficiently and not having necessarily, maybe even a human answer the phone. Maybe we're able to get that information out as they chat with us. So -- and we look at that across all the segments to try to determine the needs for each customer.

And to tack that on to a previous question, we are, I think pretty good at looking at the customer segment level and pricing to that lifetime 96 that Tricia mentioned. Obviously, we're also charged with the calendar for the organization as a 96. But we think through those average premiums you just mentioned, for the Sams and the Robinsons, but also the lifetime average -- lifetime total premium. And those Robinsons are going to stay with us a lot longer. And we build into both our acquisition pricing as well as our servicing pricing those attributes. So to the extent a customer segment costs us more for serving -- or servicing or costs us less, we have a component of our pricing that is intended to reflect that.

Thanks, Ian. Martia, we'll take the next question, please.

The next question comes from Paul Newsome from Sandler O'Neill Partners.

I was hoping you could help me kind of with a little bit of a modeling question. I'm often looking at the relationship between written premium and earned premium, Progressive tends to be a little bit more challenging in that relationship. I suspect, and here's the question, that has to do with a changing amount of 1-month, 6-month in your policies, in your book. So I guess my question is, is that true? And if it's true, could you talk about how the mix of business has changed over the last year or so? Or to between policies or different types of duration in your book -- in your Auto book?

Sure. We only write 6 or -- 6-month or 12-month policies. Commercial Lines is predominantly 12-month, our Special Lines programs are also predominantly 12-month policies. Our auto programs are still by far and away 6-month policies. However, we have opened up annual policies for auto for the Platinum agents. So that segment is growing, as we've noted. That's our intent and along with it, annual policies. But we're talking about very low single-digit percentages of our total auto policies written right now are annual. But if you have a model that is looking back and it's not matching what's happening more recently, yes, a small percentage of new business policies are now being written in annual in Auto in our Platinum agencies.

Maybe this aged me a little bit, but was -- wasn't there once the time when, that Progressive was very much a nonstandard auto writer and wrote the one-month policies? Or am I just remembering that incorrectly?

If we did write one-month policies, it's been quite some time. It's -- not to my recollection.

Yes, I don't recall that at all. We've tested the auto 12-months, and that's why we're dipping our toe in for that preferred customer in the Platinum program only for your policy to be able to match their Home policy. But I'm not familiar with the one month.

Yes, so we'll take one month of premium down but we could do that on a 6-month policy or annual policy. And the books would show the written premium for the respective policy period.

Thanks, Paul. Martia, we'll take the next caller, please.

The next question comes from Adam Klauber from William Blair.

It sounds like bundling is going well. I think you mentioned you're going to grow your Platinum agents a lot this year. Can you give us an idea of how much Platinum agents grew from the last year to this year?

Yes, I think we wrote that. I think they -- yes, we've -- how much? Double?

We've shared in Tricia's letter that the Robinsons segment within the Platinum agencies grew over 50%. So that's not the total premium for the Platinum agencies. But those agents are certainly focused on that segment. So while we are not reporting the total growth per Platinum agencies, I think that can be looked at as reflective of, generally speaking, that -- the agents' growth, that segment.

But were you asking agents in particular that actually are able to write the Platinum?

Yes, yes.

Yes, they grew -- I think they at least doubled maybe more last year and will double again in 2017. And we're pretty open on actually the number of those within our Qs and our K.

So that's in terms of agency plants.

Yes, agency number of agencies that write it. I think, actually, I'll find it.

Great. Okay. And then as far as pricing efficiency on bundling, you've just been bundling for a few years now. Where would you say you're on that curve? Are you in the third inning, fifth inning? Are you getting better by leaps and bounds? Could you just give us some idea of that improvement curve?

Yes, I think we are on first and we have a lot to go because we knew we -- on 2 sides of bundling, we -- I've talked a little bit about changing agent behavior, integrating with ASI sales team, making sure we expand their property product from ASI across the country. So that's really been our focus on the agency side. So we feel like we have -- we are rounding first at a fast pace but have a lot to do and a lot of runway to go. On the direct side, we've been working on that a lot longer and we have more -- we have ASI and some other unaffiliated partners. I feel like we're really ramping up speed on the direct side. So I talked years ago about our in-house agency that we call Progressive Advantage Agency. And it has grown substantially over the past several years. We started out 2 or 3 years ago with 25 people. We're upwards of several hundred now that are able to write Home Auto and other products. And there is -- these are your Progressive agents on the direct side. And we feel really proud about the momentum we have there. And actually, they've grown in sales over 100% this year. So a little bit different story on both, but both good stories and both have a lot of momentum.

Great. And then could you give us some idea of weather impact. I mean, it's been a very, very easy winter for most of the country. Is that helping Auto losses in January, February? And then also, San Jose, there was flooding. Is that going to be a similar event to what we saw in, maybe Houston or Baton Rouge, last year?

We feel great about our January results. Our February will wrap up this week, so you'll know soon what those results are. But we feel obviously -- we're in Cleveland, we look out and there's not much snow, so we feel pretty positive. I don't have a lot of data on San Jose but I don't believe it'll be similar to what we saw last year.

Oh, Martia, we'll take the next caller, please.

The next question comes from Brian Meredith from UBS.

A couple of questions here for you, Trish. The first, I'm just curious, you have been extending duration on your fixed income portfolio. How much more room do you have to extend duration? Should we expect that to continue to rise here? And I guess, does that imply that you think that interest rates are topping out here?

I'll go ahead and let Bill take that one. He's on the phone.

Hi. Sure. Yes, I don't think that the interest rates are topping out here. But what we've said all along is that, as rates rise, we'll gradually extend our duration as breakevens get a little bit better and we're getting paid a little bit more to take some rate risks. So as I would say, as rates rise going forward, and we expect them to continue to rise, we'd also expect to increase our duration a little bit. Now our range on duration, our guideline is 1.5 years to 5 years. I don't see us getting to 5 years any time soon. But we had 2.2 years. We have considerable room to extend our duration as we see rates go higher.

Great. And then the next question, Tricia, if I take a look at your Agency business, and I'd look at the combined ratio for '16 versus '15, and the increase you have now, I understand part of that is weather-related and catastrophe losses. But how much of it do you think is because of the growth in new business? Just kind of the new business penalty versus as you're thinking about the bundled product and more retentions? Should we expect that combined ratio to kind of keep closer to 96 going forward just because of the lifetime value of the customer?

Good question. And some of it, it's obviously the new business penalty we talked about in October both on the Commission side, and oftentimes, new business. Here, we don't know the customers well and losses will be a little bit higher. We -- as you know, we don't talk about our particular goals for combined ratio on each of our segments or on anything under the aggregate of 96. But yes, you should expect, as a company, that we will try to be at or below a 96 with all of our products combined. So we feel very positive about where we're at with agency. And we feel positive about, more importantly, having an ever evolving preferred product model for them to be able to sell, to give agents what they needed, to get us that customer that we had not have in the past. And we believe the lifetime value of customers that have more products and are more preferred and stay with us because our rates are competitive ends up having more value to Progressive.

Absolutely. And then, I know you don't talk about per product, the combined ratio targets. But as I think about it, if you're selling a bundled package policy, I would think that your homeowners targets have got to be well below a 96 given the volatility. I mean, do you think of it that way? And when you think about a particular customer, are you saying, all right, we're targeting this customer to 96, so maybe higher auto, lower homeowners?

We absolutely differentiate on what we think the customer -- each customer, the value it could bring. So looking at a bundled customer, we would look at differently than a monoline home or a monoline auto, but always getting to that 96 or lower.

Thank you. Martia, we'll take the next question, please.

[Operator Instructions] The next question comes from the line of Barry White from USAA Asset Management.

You guys have some hybrid securities coming due -- or I'm sorry, coming up for call in June. I was just wondering if you have decided on what you're planning to do with those if they're going to extend or stay put. And either way, I guess, what your -- if you're not going to do that, what your considerations are for evaluating a decision like this.

We have not decided. And Bill, do you want to give any more color on that?

Sure. Yes, we haven't decided. I'll also ask John to chime in on this question as well. It's a combination of spreads and rates. The hybrid security floats at 201 and 3/4 basis points over a 3-month LIBOR come June. And that's a relatively widespread for us and rates. And that would float, unfortunately at a -- or fortunately for us at a low rate at the moment. So that's what we're deliberating about. John, do you have more to add?

I really don't. We -- are charged with letting our owners, if we choose to redeem in June, know by mid-May, and we'll have that. But we have made no decision as of this point.

All right. Thanks, Barry. Martia, we'll take the next caller, please.

The next question comes from the line of Kai Pan from Morgan Stanley.

Just a follow-up on Brian's question on your investment portfolio. Bill, so what percentage of your fixed income portfolio mature each year? And what's the new money yield versus the maturing yields? And I just want to understand the impact from rising interest rates.

That's a good question. There is a significant part of our fixed income portfolio that matures on an annual basis and we disclose that in the A -- in the annual report. And I'll describe our new money yield -- I'll first define on new money yield as cash from operations, principal and interest paydowns as well as redeployment of the existing portfolio of cash flow. There is cash that we raise from selling some bonds and also redeploying from our short-term portfolio to a more fixed maturity portfolio. So in this fourth quarter, our new money yield was relatively low, below our book yield and actually, just a little below 2%. However, a big part of that, over $2 billion, was redeploying cash from our short-term portfolio to largely our treasury portfolio as we extend the duration. So it was higher yielding with the cash we had held, but treasury is still lower yielding than our average book yield in the portfolio.

Yes, and when I was looking through some of the documents, we talked about tripling the number of agents. So you'll read that in the letter. And then we'll -- our plan is to more than double that. So that's what we're at with Platinum agency, continue to make sure we have enough agents across the country as we continue to expand our coverage.

So that was in response -- but Adam...

Yes, that was Adam's question a little earlier.

So it looks like we have no other callers at this time. So that concludes our call today. Martia, I will turn it back over to you for the closing script, please.

That concludes the Progressive Corporation's Investor Relations conference call. An instant replay of the call will be available through Friday, March 17, by calling 1 (800) 285-0609 or it can be accessed via the Investor Relations section of Progressive's website for the next year.